report.object
AI for customer-service workflows
Expert-reviewed from public records, platform data, primary sources, and private research beyond AI search. Use /api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/sample before purchase, then download PDF, JSON, or Markdown after checkout.
page.markdown
AI for customer-service workflows.md
# AI for customer-service workflows
Customer-service AI is splitting by control rights. Helpdesk and CCaaS incumbents keep gravity when they own records and channels; specialists win when a narrow workflow needs faster action, better voice execution, or auditable outcome economics.
## Object
| field | value |
| ---------------- | -------------------------------------------- |
| report_id | rpt_ai_customer_experience_workflows_2026_05 |
| schema_version | 2026-05-28 |
| last_updated_at | 2026-06-16 |
| category | AGENTS |
| sector | Customer experience |
| region | GLOBAL |
| confidence_score | 0.88 |
| source_count | 71 |
| evidence_count | 72 |
## Endpoints
| action | endpoint |
| ------------ | ----------------------------------------------------------------------- |
| summary | /api/v1/reports/rpt_ai_customer_experience_workflows_2026_05 |
| sample | /api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/sample |
| schema | /api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/schema |
| quote | /api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/quote |
| purchase | /checkout/rpt_ai_customer_experience_workflows_2026_05?view=agent |
| versions | /api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/versions |
| changelog | /api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/changelog |
| monitor | /api/v1/monitors?report_id=rpt_ai_customer_experience_workflows_2026_05 |
| mcp_resource | colophon://reports/rpt_ai_customer_experience_workflows_2026_05/sample |
## Coverage
| field | value |
| -------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| category | AGENTS |
| sector | Customer experience |
| region | GLOBAL |
| topics | customer experience AI, customer support agents, voice agents, contact centres, CCaaS, helpdesk AI, agent assist, QA automation, workflow automation, outcome pricing, regulated CX, specialist overlays, action authority, observability, vertical support agents, availability, failover, operational resilience |
| source_classes | colophon_inference, vendor_docs, regulator, regulation, policy_framework, industry_report, government, reputable_secondary_analysis, customer_case_study, service_level_agreement, trust_page, support_documentation, status_page, pricing_documentation, pricing_page, commissioned_economic_study, status_observation, market_survey |
| update_cadence | Weekly |
## Sample JSON
```json
{
"report_id": "rpt_ai_customer_experience_workflows_2026_05",
"slug": "ai-customer-service-workflows",
"schema_version": "2026-05-28",
"title": "AI for customer-service workflows",
"abstract": "Customer-service AI is splitting by control rights. Helpdesk and CCaaS incumbents keep gravity when they own records and channels; specialists win when a narrow workflow needs faster action, better voice execution, or auditable outcome economics.",
"category": "AGENTS",
"sector": "Customer experience",
"region": "GLOBAL",
"topics": [
"customer experience AI",
"customer support agents",
"voice agents",
"contact centres",
"CCaaS",
"helpdesk AI",
"agent assist",
"QA automation",
"workflow automation",
"outcome pricing",
"regulated CX",
"specialist overlays",
"action authority",
"observability",
"vertical support agents",
"availability",
"failover",
"operational resilience"
],
"last_updated_at": "2026-06-16T00:00:00Z",
"source_count": 71,
"evidence_count": 72,
"confidence_score": 0.88,
"web_url": "/r/ai-customer-service-workflows/rpt_ai_customer_experience_workflows_2026_05",
"sample_endpoint": "/api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/sample",
"inventory_endpoint": "/api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/inventory",
"full_endpoint": "/api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/full",
"schema_endpoint": "/api/v1/reports/rpt_ai_customer_experience_workflows_2026_05/schema",
"redaction": {
"status": "sample",
"paid_sections_redacted": true,
"preview_limits": [
"summary",
"table_of_contents",
"evidence_profile"
]
},
"coverage": {
"category": "AGENTS",
"sector": "Customer experience",
"region": "GLOBAL",
"topics": [
"customer experience AI",
"customer support agents",
"voice agents",
"contact centres",
"CCaaS",
"helpdesk AI",
"agent assist",
"QA automation",
"workflow automation",
"outcome pricing",
"regulated CX",
"specialist overlays",
"action authority",
"observability",
"vertical support agents",
"availability",
"failover",
"operational resilience"
],
"source_classes": [
"colophon_inference",
"vendor_docs",
"regulator",
"regulation",
"policy_framework",
"industry_report",
"government",
"reputable_secondary_analysis",
"customer_case_study",
"service_level_agreement",
"trust_page",
"support_documentation",
"status_page",
"pricing_documentation",
"pricing_page",
"commissioned_economic_study",
"status_observation",
"market_survey"
]
},
"confidence": {
"overall_confidence": 0.88,
"source_quality": 0.87,
"source_recency": 0.94,
"source_density": 0.93,
"claim_verifiability": 0.84,
"geographic_coverage": 0.82,
"known_gaps": [
"Independent CX-agent benchmarks remain thin for containment, false resolution, repeat contacts, latency, escalation, action reliability, billing events and workflow-specific human baselines.",
"Pricing units vary across outcome, resolution, automated interaction, minute, message, seat, platform, implementation and vertical package models.",
"Most public deployment metrics are vendor-published and do not expose denominator, sample, control group, measurement window or audit method.",
"Availability evidence is fragmented across contractual SLAs, marketing claims and component incident records; full-route dependency failure, SLA-breach documentation and business-impact recovery usually need buyer-specific outage testing.",
"Connector depth, writeback authority, rollback, retention, subprocessor and model-provider terms often require NDA or contract review."
]
},
"document": {
"object": "report_document",
"schemaVersion": "2026-05-28",
"title": "AI for customer-service workflows",
"subtitle": "Action rights, handoff discipline and outcome proof beat the best answer bot.",
"abstract": "Customer-service AI is splitting by control rights. Helpdesk and CCaaS incumbents keep gravity when they own records and channels; specialists win when a narrow workflow needs faster action, better voice execution, or auditable outcome economics.",
"category": "AGENTS",
"sector": "Customer experience",
"region": "GLOBAL",
"topics": [
"customer experience AI",
"customer support agents",
"voice agents",
"contact centres",
"CCaaS",
"helpdesk AI",
"agent assist",
"QA automation",
"workflow automation",
"outcome pricing",
"regulated CX",
"specialist overlays",
"action authority",
"observability",
"vertical support agents",
"availability",
"failover",
"operational resilience"
],
"sources": [
{
"id": "S1",
"label": "Colophon synthesis - vendor-directory coverage method",
"publisher": "Colophon",
"sourceClass": "colophon_inference",
"accessedAt": "2026-06-16",
"status": "partial",
"access": "sample",
"caveat": "Colophon synthesis row derived from the public vendor, pricing, regulator, status, SLA and case-study sources rendered in this ledger; not an original external source.",
"transformChain": []
},
{
"id": "S2",
"label": "Colophon synthesis - capability and proof map",
"publisher": "Colophon",
"sourceClass": "colophon_inference",
"accessedAt": "2026-06-16",
"status": "partial",
"access": "sample",
"caveat": "Colophon synthesis row derived from the public vendor, pricing, regulator, status, SLA and case-study sources rendered in this ledger; not an original external source.",
"transformChain": []
},
{
"id": "S3",
"label": "Colophon synthesis - market-structure method",
"publisher": "Colophon",
"sourceClass": "colophon_inference",
"accessedAt": "2026-06-16",
"status": "partial",
"access": "sample",
"caveat": "Colophon synthesis row derived from the public vendor, pricing, regulator, status, SLA and case-study sources rendered in this ledger; not an original external source.",
"transformChain": []
},
{
"id": "S4",
"label": "Sierra Agent Platform product and customer evidence",
"publisher": "Sierra",
"sourceClass": "vendor_docs",
"accessedAt": "2026-06-16",
"status": "verified",
"access": "sample",
"transformChain": [],
"url": "https://sierra.ai/"
},
{
"id": "S5",
"label": "Decagon AI Agents product and customer evidence",
"publisher": "Decagon",
"sourceClass": "vendor_docs",
"accessedAt": "2026-06-16",
"status": "verified",
"access": "sample",
"transformChain": [],
"url": "https://decagon.ai/"
},
{
"id": "S6",
"label": "Intercom Fin product and customer evidence",
"publisher": "Intercom",
"sourceClass": "vendor_docs",
"accessedAt": "2026-06-16",
"status": "verified",
"access": "sample",
"transformChain": [],
"url": "https://www.intercom.com/fin"
},
{
"id": "S10",
"label": "Vapi product and customer evidence",
"publisher": "Vapi",
"sourceClass": "vendor_docs",
"accessedAt": "2026-06-16",
"status": "verified",
"access": "sample",
"transformChain": [],
"url": "https://vapi.ai/platform"
}
],
"evidence": [
{
"id": "E1",
"sourceId": "S1",
"claimIds": [
"C1"
],
"summary": "Public signal from S4-S20, S30-S39, S40-S45 and S58-S67: vendor evidence spans omnichannel automation, vertical agents, voice platforms, CRM/helpdesk, CCaaS, QA/analytics, speech, telephony and public-safety boundary examples. Report inference: the market needs lane segmentation rather than one chatbot category.",
"status": "verified",
"confidence": 0.94,
"access": "sample"
},
{
"id": "E2",
"sourceId": "S2",
"claimIds": [],
"summary": "Public signal from S4-S20, S30-S39, S40-S45 and S58-S67: product pages and customer evidence show stronger product/deployment proof than independent benchmark proof. Report inference: public evidence can route diligence but cannot approve performance.",
"status": "verified",
"confidence": 0.94,
"access": "sample"
},
{
"id": "E3",
"sourceId": "S3",
"claimIds": [],
"summary": "Public signal from S4-S24 and S30-S39: vendor and governance evidence repeatedly turns on data access, action rights, channel control, escalation and audit. Report inference: CX AI is a workflow-execution layer, not only an answer surface.",
"status": "verified",
"confidence": 0.92,
"access": "sample"
},
{
"id": "E4",
"sourceId": "S4",
"claimIds": [
"C2"
],
"summary": "Sierra evidence supports an enterprise agent operating system with secure system actions, voice payments, memory, decisioning, simulations, observability, outcome pricing, and named customer stories; pricing and independent evals remain limited.",
"status": "verified",
"confidence": 0.9,
"access": "sample"
},
{
"id": "E5",
"sourceId": "S5",
"claimIds": [
"C2"
],
"summary": "Decagon evidence supports AOPs, integrations, tests, Watchtower QA, proactive agents, chat/email/voice/SMS channels, and vendor-published Chime, Rippling, ClassPass, Rituals, and Noom outcomes.",
"status": "verified",
"confidence": 0.89,
"access": "sample"
},
{
"id": "E6",
"sourceId": "S6",
"claimIds": [
"C2"
],
"summary": "Fin evidence shows grounded answers, Data Connectors, API actions, procedures, omnichannel handoff, public outcome pricing mechanics, and case studies; benchmark claims remain vendor-run or opaque.",
"status": "verified",
"confidence": 0.9,
"access": "sample"
},
{
"id": "E10",
"sourceId": "S10",
"claimIds": [
"C3"
],
"summary": "Vapi evidence supports a developer voice-agent control layer with assistants, squads, telephony/WebRTC/WebSocket/SIP, tool calling, provider routing, logs, data-flow controls, per-minute pricing, and named production cases.",
"status": "verified",
"confidence": 0.88,
"access": "sample"
}
],
"claims": [
{
"id": "C1",
"statement": "Customer-service AI is not one chatbot category; the market spans at least eight stack roles.",
"confidence": 0.94,
"evidenceIds": [
"E1",
"E4",
"E5",
"E6",
"E10"
],
"access": "sample"
},
{
"id": "C2",
"statement": "The strongest differentiator is controlled workflow execution: data grounding, action permissioning, testing, escalation, and audit.",
"confidence": 0.92,
"evidenceIds": [
"E4",
"E5",
"E6"
],
"access": "sample"
},
{
"id": "C3",
"statement": "Voice is becoming a premium CX frontier because phone workflows carry high labour cost and high failure cost.",
"confidence": 0.88,
"evidenceIds": [
"E10"
],
"access": "sample"
},
{
"id": "C12",
"statement": "The buyer route should be selected by customer data control, action permissions, channel interface, escalation path, and audit loop.",
"confidence": 0.94,
"evidenceIds": [
"E4",
"E5",
"E6",
"E10"
],
"access": "sample"
}
],
"colophon": {
"checkedAt": "2026-06-16",
"latestRevisionAt": "2026-06-16T00:00:00Z",
"reviewStandard": "expert_human_reviewed",
"kinetics": {
"speed": "fast",
"impact": 0.9,
"momentum": 0.91,
"businessImpact": "CX AI is moving from self-service answer bots into route-approved agents that touch orders, payments, appointments, cases, phone queues, QA records, and regulated customer data.",
"rateOfChange": "Native helpdesk-agent releases, outcome-billing definitions, AI disclosure enforcement and voice reliability evidence are moving fast enough to reopen shortlists within a quarter.",
"from": "chatbot deflection, IVR containment and agent-assist pilots",
"toward": "route-approved workflow execution with action gates, observable handoff, QA loops, audit and billing controls"
},
"watch": [
"2 August 2026: EU AI Act transparency obligations for customer-facing AI interactions apply; AI disclosure evidence could become a launch gate for chat and voice routes. [E23, E28]",
"FCC TCPA or robocall enforcement releases naming synthetic-voice disclosure, identity statement, consent, opt-out or DNC failures; outbound phone-agent authority could narrow. [E21]",
"CFPB enforcement action, circular or report naming AI chatbot complaint mishandling in bank or fintech servicing; financial-services support agents could need tighter human escalation and complaint evidence. [E22]",
"Zendesk or Salesforce release notes adding native autonomous writeback in support workflows; helpdesk and CRM-native routes could move ahead of specialist overlays.",
"Fin or Zendesk outcome-pricing pages changing definitions of resolution, containment, assisted escalation or billable handoff; accepted-resolution cost models could change. [E61, E62]",
"Vapi public status page publishes a named multi-hour API, database, telephony, speech or webhook incident postmortem; developer-voice availability and failover requirements could tighten. [E52, E53]"
],
"gapAnalysis": "Independent end-to-end CX-agent benchmarks are still thin: public evidence rarely links action reliability, false resolution, repeat contact, escalation quality, latency, availability, dependency outages, consent, billing events, raw logs, and workflow-specific human baselines."
},
"body": {
"sections": [
{
"id": "executive-summary",
"title": "Executive summary",
"access": "sample",
"dek": "The answer bot is becoming the action layer.",
"claimIds": [
"C1",
"C2",
"C3",
"C12"
],
"evidenceIds": [
"E1",
"E2",
"E3",
"E4",
"E5",
"E6",
"E10"
],
"blocks": [
{
"id": "executive-summary-narrative",
"evidenceIds": [
"E1",
"E2",
"E3",
"E10"
],
"type": "section_narrative",
"lead": "The best CX AI product is the one allowed to finish the workflow and prove what happened.",
"body": [
"The market is a stack fight among CRM and helpdesk platforms, CCaaS incumbents, AI-native agent platforms, vertical phone agents, developer voice infrastructure, and QA or agent-assist layers. Ranking vendors by automation rate would flatten the useful differences: answering a FAQ, changing an order, booking an appointment and handling a regulated call do not carry the same cost, risk or proof burden. [E1, E2]",
"The buying question is permission. Which vendor can see the customer record, use the right knowledge, take the permitted action, hand off at the right time, and leave an audit trail that operations, security and compliance can inspect? [E3] Voice is where that test gets harder: latency, interruption, consent, recording, disclosure, fraud, payments, PHI and escalation collide in real time. [E10, E21]"
],
"detail": [
"Treat vendor-published ROI and containment metrics as deployment signals, not comparable performance evidence. Most public cases omit sample design, denominator, repeat-contact treatment, measurement window, and independent audit. [E2]"
]
},
{
"id": "executive-decision-table",
"title": "What changes the shortlist",
"evidenceIds": [
"E2",
"E3",
"E4",
"E6",
"E10"
],
"type": "table",
"columns": [
{
"key": "question",
"label": "question",
"align": "left"
},
{
"key": "why",
"label": "why it matters",
"align": "left"
},
{
"key": "buyerMove",
"label": "buyer move",
"align": "left"
}
],
"rows": [
{
"question": "Where is the customer record?",
"why": "CRM/helpdesk incumbents gain leverage when identity, history, permissions and case state already live there.",
"buyerMove": "Test the native agent first; use an overlay only if the workflow crosses systems or native permissions are too shallow."
},
{
"question": "What can the agent do?",
"why": "Answers are cheap; refunds, appointments, payments, booking, routing and account changes need permission and rollback.",
"buyerMove": "Separate answer-only, proposed action, writeback, money movement, outbound voice and regulated-record authority. Approve each tier separately."
},
{
"question": "Which channel carries the failure cost?",
"why": "Voice and regulated channels require lower latency, stronger disclosure and cleaner escalation.",
"buyerMove": "Run phone and outbound approval separately, with latency, disclosure, transfer-failure and consent tests."
},
{
"question": "How is success billed?",
"why": "Outcome pricing can hide repeated contacts, partial resolutions and assisted handoffs.",
"buyerMove": "Reconcile invoices to transcripts, tickets, handoffs, repeat contacts and human interventions."
}
]
},
{
"id": "demo-risk-bridge",
"evidenceIds": [
"E2",
"E6",
"E10"
],
"type": "paragraph",
"text": "Once those questions choose the route, the next gate is what the demo can conceal before approval.",
"variant": "body"
},
{
"id": "executive-traps-table",
"title": "What the demo hides",
"evidenceIds": [
"E2",
"E6",
"E10"
],
"type": "table",
"columns": [
{
"key": "trap",
"label": "trap",
"align": "left"
},
{
"key": "whatItHides",
"label": "what it hides",
"align": "left"
},
{
"key": "betterTest",
"label": "buyer test before approval",
"align": "left"
}
],
"rows": [
{
"trap": "High containment",
"whatItHides": "False resolution, repeat contact and quiet transfer failure.",
"betterTest": "Count accepted resolutions only after 7/14/30-day follow-up windows, excluding reopenings, transfers and same-issue repeats across channels."
},
{
"trap": "Integration logos",
"whatItHides": "Read access without writeback, rollback, idempotency or audit export.",
"betterTest": "Execute the target write actions in a sandbox; verify writeback, rollback, idempotency, permissions and audit logs."
},
{
"trap": "Human-sounding voice",
"whatItHides": "Latency, interruption, consent, payment and escalation failure under real calls.",
"betterTest": "Test noisy, accented, emotional and interrupted calls that include consent, payment and escalation steps before approval."
},
{
"trap": "Outcome pricing",
"whatItHides": "Billable handoffs, assisted resolutions, partial actions and repeats.",
"betterTest": "Reconcile invoices to transcripts, tickets and human interventions; exclude handoffs, assisted resolutions, partial actions and repeat contacts from outcomes."
}
]
},
{
"id": "route-definition-callout",
"title": "Route definition",
"evidenceIds": [
"E1",
"E2",
"E3",
"E4",
"E5",
"E6",
"E10"
],
"type": "callout",
"tone": "neutral",
"body": "Route means the approved path from work to accepted output: the source set, data boundary, model or vendor channel, tools, reviewer, evidence record and reapproval trigger that make an AI result usable."
}
]
},
{
"id": "decision-map",
"title": "Decision frame",
"access": "sample",
"dek": "What changed is permission: the buyer now approves routes, not chatbots.",
"claimIds": [
"C2",
"C12"
],
"evidenceIds": [
"E2",
"E3",
"E4",
"E5",
"E6",
"E10"
],
"blocks": [
{
"id": "decision-map-narrative",
"evidenceIds": [
"E3",
"E10"
],
"type": "section_narrative",
"lead": "Segment the market by the system that owns the work: customer record, channel, workflow engine, vertical system, developer stack, or QA loop.",
"body": [
"What changed for buying is that customer-service AI now asks for permission to act, bill, hand off and leave evidence. If support already runs in Zendesk or Fin, test the native route first; Salesforce and Gorgias follow the same logic where CRM or commerce context dominates. [E7, E6] [E30, E8] If phone routing, workforce, recording and supervisor workflow dominate, start with CCaaS. If the work crosses systems, test an AI-native overlay.",
"Use developer voice platforms when the buyer has the engineering team to finish the product. Vapi and Retell supply programmable voice primitives. [E10, E12] Bland supports voice-agent pathways, tools and guardrails. [E11] The buyer still owns escalation, tool-call permissions, consent, retention and QA. Treat QA and agent-assist tools as supervision layers: they improve the reps and bots already in production; they are not failed autonomous agents."
]
},
{
"id": "cx-routing-tree",
"title": "Control point chooses the first lane",
"evidenceIds": [
"E3",
"E6",
"E10"
],
"type": "decision_tree",
"lead": "Start with the control point, then test the agent.",
"steps": [
{
"prompt": "Is the target workflow inside an existing helpdesk, CRM or commerce platform?",
"tone": "info",
"branches": [
{
"label": "Yes",
"outcome": "Test the native agent first: Fin and Zendesk for helpdesk-native routes, or Salesforce and Gorgias where CRM or commerce context dominates. [E6, E7] [E30, E8]",
"tone": "success"
},
{
"label": "No",
"outcome": "Continue to channel and workflow ownership.",
"next": "Q2",
"tone": "monitor"
}
]
},
{
"prompt": "Is phone routing, recording, workforce, QA and supervisor workflow central?",
"tone": "warning",
"branches": [
{
"label": "Yes",
"outcome": "Start with sourced CCaaS or contact-centre AI examples: Amazon Connect and Genesys, then Five9 and Talkdesk. [E9, E31] [E32, E33]",
"tone": "info"
},
{
"label": "No",
"outcome": "Continue to workflow depth.",
"next": "Q3",
"tone": "monitor"
}
]
},
{
"prompt": "Is the work industry-specific enough to pre-package the workflow?",
"tone": "success",
"branches": [
{
"label": "Yes",
"outcome": "Use the market map to route the lane, then evaluate vertical, healthcare or financial-services work under its own operating standard. [E13, E14]",
"tone": "success"
},
{
"label": "No",
"outcome": "Use an AI-native agent platform or developer voice stack, but require private workflow evals.",
"tone": "warning"
}
]
}
]
},
{
"id": "decision-map-section-callout",
"title": "Route by control",
"evidenceIds": [
"E2",
"E3",
"E4",
"E5",
"E6",
"E10"
],
"type": "callout",
"tone": "info",
"body": "Answer quality is a weak gate for order changes, payments, appointments and case closure; require separate action authority."
}
]
},
{
"id": "incumbents-vs-specialists",
"title": "Incumbents vs specialists",
"access": "locked",
"dek": "Procurement gravity meets workflow focus.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "incumbents-vs-specialists-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "workflow-proof-map",
"title": "Workflow proof map",
"access": "locked",
"dek": "Signature proof maps where public evidence can and cannot prove workflow control; logo counts are too thin.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "workflow-proof-map-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "metrics-and-benchmarks",
"title": "Metrics and benchmarks",
"access": "locked",
"dek": "The public numbers are useful only after the buyer rewrites the denominator.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "metrics-and-benchmarks-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "use-case-routing",
"title": "Use-case routing",
"access": "locked",
"dek": "The right lane changes with the permitted action.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "use-case-routing-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "voice-and-verticalization",
"title": "Voice and vertical boundaries",
"access": "locked",
"dek": "The phone is where generic automation claims get expensive, but sector approval belongs to specialist operating standards.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "voice-and-verticalization-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "data-control-plane",
"title": "Data and control planes",
"access": "locked",
"dek": "Connector lists matter only after the buyer can see what the agent may do.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "data-control-plane-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "availability-failover",
"title": "Availability and failover",
"access": "locked",
"dek": "Availability decides how much of the customer path an AI agent may own.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "availability-failover-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "pricing-and-economics",
"title": "Pricing and economics",
"access": "locked",
"dek": "The hard unit is a safely completed workflow, not a message.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "pricing-and-economics-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "deployment-evidence",
"title": "Deployment evidence",
"access": "locked",
"dek": "Named customer stories are useful, but not portable benchmarks.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "deployment-evidence-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "evaluation-controls",
"title": "Evaluation and controls",
"access": "locked",
"dek": "Controls must follow the workflow, not the vendor category.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "evaluation-controls-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "operating-model",
"title": "Operating model",
"access": "locked",
"dek": "Name control owners before vendor management.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "operating-model-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "known-unknowns",
"title": "Known unknowns",
"access": "locked",
"dek": "Evidence gaps change the buying decision.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "known-unknowns-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "recommended-next-steps",
"title": "Recommended next steps",
"access": "locked",
"dek": "A buyer should buy the workflow route, not the demo.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "recommended-next-steps-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "vendor-directory",
"title": "Vendor appendix",
"access": "locked",
"dek": "A post-decision reference list, not another provider landscape.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "vendor-directory-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "source-ledger",
"title": "Source ledger",
"access": "locked",
"dek": "Primary sources, synthesis rows and claim support behind the analysis.",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "source-ledger-locked",
"title": "Locked section",
"evidenceIds": [],
"type": "callout",
"tone": "locked",
"body": "Purchase or subscribe to unlock this section."
}
]
},
{
"id": "glossary",
"title": "Glossary",
"access": "sample",
"claimIds": [],
"evidenceIds": [],
"blocks": [
{
"id": "glossary-table",
"title": "Terms used in this report",
"caption": "Definitions are reader aids, not separate evidence claims.",
"evidenceIds": [],
"type": "table",
"columns": [
{
"key": "term",
"label": "Term",
"align": "left"
},
{
"key": "meaning",
"label": "Plain meaning",
"align": "left"
}
],
"rows": [
{
"term": "Route",
"meaning": "The approved path from work to accepted output: source set, data boundary, model or vendor channel, tools, reviewer, evidence record and reapproval trigger."
},
{
"term": "API",
"meaning": "Application programming interface: a governed software connection used by an app or workflow to call a model or service."
},
{
"term": "LLM",
"meaning": "Large language model: a model trained to read and generate text-like tokens."
},
{
"term": "Token",
"meaning": "A chunk of text used for billing, context windows and rate limits. It is usually smaller than a word, not a page count."
},
{
"term": "PayGo",
"meaning": "Pay as you go: shared cloud capacity billed by usage rather than pre-reserved throughput."
},
{
"term": "SLA",
"meaning": "Service-level agreement: a contractual availability or support commitment, not observed uptime history."
}
]
}
]
}
]
}
},
"gap_analysis": "Independent end-to-end CX-agent benchmarks are still thin: public evidence rarely links action reliability, false resolution, repeat contact, escalation quality, latency, availability, dependency outages, consent, billing events, raw logs, and workflow-specific human baselines.",
"kinetics": {
"speed": "fast",
"impact": 0.9,
"momentum": 0.91,
"businessImpact": "CX AI is moving from self-service answer bots into route-approved agents that touch orders, payments, appointments, cases, phone queues, QA records, and regulated customer data.",
"rateOfChange": "Native helpdesk-agent releases, outcome-billing definitions, AI disclosure enforcement and voice reliability evidence are moving fast enough to reopen shortlists within a quarter.",
"from": "chatbot deflection, IVR containment and agent-assist pilots",
"toward": "route-approved workflow execution with action gates, observable handoff, QA loops, audit and billing controls"
},
"watch_signals": [
"2 August 2026: EU AI Act transparency obligations for customer-facing AI interactions apply; AI disclosure evidence could become a launch gate for chat and voice routes. [E23, E28]",
"FCC TCPA or robocall enforcement releases naming synthetic-voice disclosure, identity statement, consent, opt-out or DNC failures; outbound phone-agent authority could narrow. [E21]",
"CFPB enforcement action, circular or report naming AI chatbot complaint mishandling in bank or fintech servicing; financial-services support agents could need tighter human escalation and complaint evidence. [E22]",
"Zendesk or Salesforce release notes adding native autonomous writeback in support workflows; helpdesk and CRM-native routes could move ahead of specialist overlays.",
"Fin or Zendesk outcome-pricing pages changing definitions of resolution, containment, assisted escalation or billable handoff; accepted-resolution cost models could change. [E61, E62]",
"Vapi public status page publishes a named multi-hour API, database, telephony, speech or webhook incident postmortem; developer-voice availability and failover requirements could tighten. [E52, E53]"
],
"sample_findings": [
"Customer-service AI is moving from chatbot deflection to governed workflow execution."
]
}
```